All dialects: resyllabification

< Previous | Next >

princeipeazul

Senior Member
Filipino
According to the book Colloquial Arabic of the Gulf: 2nd edition, there is what they call a syllable rearrangement when an object pronoun is attached to a verb in the imperfect verb. Thus تشربه is pronounced as /təˈʃɪɾ.ba/ instead of /ˈtɪʃ.ɾa.ba/.

The same thing tends to happen with the feminine and plural of the imperative. Thus اشربي /ˈɪʃ.ɾa.bɪ/ (< ishrab + i) is often heard as /ˈʃɪɾ.bɪ/, and اشربوا /ˈɪʃ.ɾa.bu/ (< ishrab + u) as /ˈʃɪɾ.bu/. One also hears /ˈkɪt.bɪ/ instead of /ˈɪk.tɪ.bɪ/ إكتبي, etc. And finally يشربه tends to become /jəˈʃɪɾ.ba/ instead of /ˈjɪʃ.ɾa.ba/ (please take note of the change in stress)

I'm interested to know if this syllable-arrangement is exclusive to Gulf Arabic dialects or does it also occur to other dialects especially to Najdi?
 
Last edited:
  • Hemza

    Senior Member
    French, Mor/Hijz Arabic (heritage)
    Hello,

    This stress is the usual one in the Maghreb with a slight difference for some Maghrebi dialects which pronounce / jiˈʃəɾ.bah (buh) / or /ˈʃuɾ.bɪ / for instance. More broadly, the different stress (mainly on words, not really on verbs) is one of the reasons which make these dialects hard to understand for Egyptian/Sudanese and Levantine Arabic speakers as the pronunciation sounds unusual to them.
     
    Last edited:

    fenakhay

    Senior Member
    French (France) / Arabic (Morocco)
    In my dialect, it is /ˈʃuɾ.bi/ but for masculine, it is /ˈʃɾu.b/.
     

    wriight

    Senior Member
    English (US) / Arabic (Lebanon)
    or does it also occur to other dialects especially to Najdi?
    @Wadi Hanifa is probably the best-equipped to explain how exactly this resyllabification manifests in Najdi, but I'm fairly certain that some form of it occurs in every Arabic dialect today.

    I can explain what motivates it in North Levantine, where it does also occur, although not to the extent indicated by those Gulf examples. But first, it's important to note that this resyllabification isn't specifically the result of adding an object pronoun -- instead, it's just the result of the word being lengthened by a syllable (which adding an object pronoun does do, but it's not the only way to get the same end result is what I'm saying). But sure, to keep things simple we can think about it in terms of adding a suffix.

    If the suffix being added is vowel-initial, then it snags the last consonant of the base word into a new syllable, and that's basically all that this phenomenon boils down to. The reason that causes resyllabification is that there's a categorical prohibition on the structure [stressed heavy syllable] [unstressed open syllable ending in kasra] [unstressed syllable], which often arises from the addition of a vowel-initial suffix to a sequence of [stressed syllable] [unstressed, closed syllable containing kasra]. That "snagging" opens the second syllable and adds a third unstressed one to the end, and as soon as that happens, the "base" word must resyllabify in order to stop the 'forbidden structure' from surfacing.

    (FWIW: the loanword kaamera "camera" is in plain violation of this rule. There are a number of possible explanations, although they're off topic for this thread... in any case, we can ignore it here, because the rule holds in all other scenarios.)

    For example, let's take byiktbu "they write", a conjugated form of the verb katab "to write". Underlyingly, it's composed of byiktib + -u (bolding for stress), but there's a problem: just sticking those two parts together gets us *byiktibu, and if we syllabify that, we find that it contains the forbidden sequence! That is, it breaks down into byik (stressed heavy syllable) + ti (unstressed open syllable ending in kasra) + bu (unstressed syllable). So the word is forced to resyllabify in order to avoid this, ultimately yielding byiktbu or byiketbu on the surface.
    In Egyptian Arabic, this particular case isn't a problem, thanks to stress being fixed on the penultimate syllable: Egyptian biyiktib + -u for "they write" simply results in biyiktibu, where the kasra-final syllable's being stressed avoids any problems it'd otherwise cause.

    Back to North Levantine, we can use this same rule to explain why resyllabification doesn't happen with a verb on the pattern of "yif3al", only "yif3il/yif3ul" (the latter two are the same thing, as "yif3ul" becomes "yif3il" when suffixed to). Looking at the verb byishrabu "they drink", a conjugated form of the verb shirib "to drink", we can see that it too is composed of byishrab + -u. But the difference is that, this time, "just sticking those two parts together" gives us no unstressed syllable-final kasra at all, meaning that the naive result byishrabu actually has no problem being the actual realization of this verb.

    And one more example of resyllabification not happening. The expected result of "he writes it.f" in Gulf Arabic, going by your examples, is /jəˈkɪt.ba/ (edit: this was wrong!)-- but the equivalent in most of North Levantine is byiktiba, not byiktba as might be suggested from the above. This is because the 3sg.f and 3pl suffixes, -a and -un, still come with vestiges of the fact that they used to be -ha and -hun: with the consonant at the beginning, we can clearly see that the result of byiktib + -ha is byiktibha, where the 'forbidden sequence' isn't even close to appearing, and so even with the lost -h- the stress pattern remains in byiktiba sans restructuring. I believe areas of Lebanon do fully regularize these two suffixes, though, resulting in restructured byiktba as would naively be expected in other regions from byiktib + -a.
    (The "vestiges of the fact that they used to be -ha and -hun" thing is also why your /jəˈʃɪɾ.ba/ changes its stress to the penultimate syllable (edit: also wrong!), but Gulf Arabic resyllabification clearly operates under different rules than the ones I'm describing for Levantine, and it looks like the restructuring isn't even influenced by the kasra-final-syllable thing.)

    The reason I'm singling out North Levantine specifically is that this list of conjugations from elroy for Palestinian Arabic caught my eye the other day:
    batalwen, bittalwen, bittalwini, bitalwen, bittalwen, mintalwen, bittalwinu, bitalwinu
    talwanet, talwanet, talwanti, talwan, talwanat, talwanna, talwantu, talwanu
    Emphasis mine. In Lebanese and probably North Levantine in general, the expected forms of the two bolded ones would be bit.ta.lw.nu and bi.ta.lw.nu, both resyllabified. (The "lw" syllable, pronounced lu, is the surface realization of *liw, which might more-clearly show where the word resyllabified.) I'm not sure in what respect South Levantine differs with regard to its prohibition of the 'forbidden sequence', but it means I can't explain "Levantine" behavior in general with the description above.
     
    Last edited:

    Hemza

    Senior Member
    French, Mor/Hijz Arabic (heritage)
    And one more example of restructuring not happening. The expected result of "he writes it.f" in Gulf Arabic, going by your examples, is /jəˈkɪt.ba/ -- but the equivalent in most of North Levantine is byiktiba, not byiktba as might be suggested from the above! This is because the 3sg.f and 3pl suffixes, -a and -un, still come with vestiges of the fact that they used to be -ha and -hun.
    Actually there is no feminine here. It is يكتبه and not يكتبها, not to be confused by how in Lebanon you pronounce the suffixed ها. So it is "yekitba(h)" implying a masculine suffix so "he writes it.m"
     

    elroy

    Imperfect Mod
    US English, Palestinian Arabic bilingual
    In Lebanese and probably North Levantine in general, the expected forms of the two bolded ones would be bit.ta.lw.nu and bi.ta.lw.nu, both subjected to restructuring.
    I’m not sure “talwan” is used in Galilean. If it is, then the Galilean forms would in fact be “bittaliwnu” and “bitaliwnu.”

    By the way, the term for this is “resyllabification” (not “restructuring”).
     

    wriight

    Senior Member
    English (US) / Arabic (Lebanon)
    Actually there is no feminine here. It is يكتبه and not يكتبها, not to be confused by how in Lebanon you pronounce the suffixed ها. So it is "yekitba(h)" implying a masculine suffix so "he writes it.m"
    I see (I thought this was only an Iraqi feature, didn't know about Gulf). So the feminine remains -ha? And, since it looks like -a(h) causes stress to shift to the penultimate syllable, do second-person -ek (ك) and -eč (چ) cause it to as well?
    By the way, the term for this is “resyllabification” (not “restructuring”).
    That's much cleaner, edited. I was misquoting the original post's "rearrangement" :oops:
     
    Last edited:

    Hemza

    Senior Member
    French, Mor/Hijz Arabic (heritage)
    I see (I thought this was only an Iraqi feature, didn't know about Gulf). So the feminine remains -ha? And, since it looks like -a(h) causes stress to shift to the penultimate syllable, do second-person -ek (ك) and -eč (چ) cause it to as well?
    This "ah" isn't responsible of the stress shift I think. This ending is not specifically Gulfian or Iraqi but also exists in Najd (except one place :D القسيم) amongst bedouin dialect speakers of the Levant as well as in Sinai, Western Egypt, Libya, Chad, I think few places in Tunisia, some places in Eastern/central Algeria as well as Western Algeria and Eastern Morocco. So عنده for instance, is pronounced "3inda(h)/3enda(h)" instead of the "u" ending while the feminine remains "ha".

    As for your question, I'm unable to reply :(.
     
    Last edited:

    analeeh

    Senior Member
    English - UK
    If I'm correct, the stress shift is probably a product of the gahawa phenomenon which appears all over this area - insertion of an epenthetic vowel followed by stress shift to the vowel. But I don't know, and I'm surprised that it seems to happen in some places rather than others.
     

    princeipeazul

    Senior Member
    Filipino
    If I'm correct, the stress shift is probably a product of the gahawa phenomenon which appears all over this area - insertion of an epenthetic vowel followed by stress shift to the vowel. But I don't know, and I'm surprised that it seems to happen in some places rather than others.
    I'm getting your point. Do you think it's due to the fact that these dialects favor initial consonant clusters? That's why I think تشربه is pronounced as /ˈtʃɪɾ.bə/ I'm not sure, but it appears to be like that...
     
    Last edited:

    Hemza

    Senior Member
    French, Mor/Hijz Arabic (heritage)
    I'm getting your point. Do you think it's due to the fact that these dialects favor initial consonant clusters? That's why I think تشربه is pronounced as /ˈtʃɪɾ.bə/ I'm not sure, but it appears to be like that...
    This is highly plausible since this stress pattern mainly (or always?) occurs amongst dialects use consonnant clusters at the beginning of verbs. But let's wait for others confirming or infirming it since I'm not sure at all of my statement.
     
    < Previous | Next >
    Top